Effect of Hepatitis G Virus Infection on Chronic Hepatitis C

Methods

Patients

We enrolled 189 patients with chronic hepatitis C who were seen at Shinshu University Hospital, Matsumoto, Japan (122 men and 67 women; mean age, 51.1 ± 11.0 years). All patients were positive for antibody to HCV according to a second-generation assay and were negative for hepatitis B virus (HBV) surface antigen and antibody to human immunodeficiency virus. No patients had hepatocellular carcinoma or a history of alcohol intake exceeding 80 g/d. Of the 189 patients, 101 (66 men and 35 women; mean age, 50.0 ± 11.2 years) had been retrospectively selected so that our sample would include all patients who had received a single course of interferon-α therapy between October 1991 and December 1993. The remaining 88 (47%) patients (56 men and 32 women; mean age, 52.5 ± 10.7 years) were consecutively selected from patients with chronic hepatitis C who had been followed for more than 1 year and had had liver biopsy within the same period but had not received antiviral therapy.

Interferon-α 2a had been administered at a dosage of 9 million U daily for 2 weeks, followed by 9 million U three times a week for 22 weeks (total dose, 720 million U). Treated patients had histologic examination within the 6 months before interferon-α therapy was initiated and were followed for at least 6 months after therapy was completed. For all patients, serum samples were obtained at the time of liver biopsy and were stored at −70°C until testing. For patients receiving interferon-α therapy, serum samples were also collected just before therapy began, just after therapy was completed, and 6 months after therapy was completed.

Serum alanine aminotransferase levels (normal range, 5 to 45 IU/L) were measured at liver biopsy and, in patients receiving interferon-α therapy, at least once every 4 weeks during therapy and follow-up. The grade (extent of hepatic inflammation and hepatocellular destruction) and stage (degree of fibrosis) of liver histologic findings [7] were judged by three authors; the final diagnosis was established by consensus. Investigators involved in separate portions of the study were blinded to the results of other portions.

Serologic Markers of and Molecular Assays for Hepatitis C Virus RNA

Levels of HCV antibody, HBV surface antigen, and antibody to human immunodeficiency virus were measured by using commercially available second-generation enzyme-linked immunosorbent assays (Abbott Laboratories, North Chicago, Illinois). Serum levels of HCV RNA were measured by using nested reverse-transcription polymerase chain reaction (PCR) with primers in the 5′ noncoding region [8] and were quantified by using the branched-DNA signal amplification assay [9, 10]. The detection limit of the branched-DNA assay was set at 10⁵.7 equivalents/mL. Hepatitis C virus genotypes were tested by nested PCR using genotype-specific primers of core region [11] and were categorized according to the classification system of Simmonds and colleagues [12].

Serum Levels of Hepatitis G Virus RNA

Levels of HGV RNA in serum were measured by using reverse-transcription PCR as described elsewhere [5]. Total nucleic acids were extracted from 50 mL of serum. After reverse transcription, 45 PCR cycles (for detection) or 35 PCR cycles (for quantitation) were done with primers in the putative NS5 region of the HGV RNA genome. The PCR products were analyzed by dot-blot hybridization with a ³²P-labeled oligonucleotide probe. The sensitivity of this assay system was 20 RNA copies/mL of starting serum. Quantitative measurement of HGV RNA levels was done using standards of known HGV RNA levels.

Statistical Analysis

Statistical analyses were done using the Student t-test, the Mann-Whitney U test, the Wilcoxon rank-sum test, the chi-square test, the Fisher exact test, and the Somers D test. A P value of 0.05 or less indicated statistical significance.

Results

Clinical Features and Hepatitis C Virus Markers

Twenty-one (11.1%) of the 189 enrolled patients were positive for HGV RNA. The rate of detection of HGV RNA was similar in the subgroup of 88 untreated patients (12.5%; 11 of 88) and the 101 patients who received interferon-α therapy (9.9%; 10 of 101). Mean age and the number of men and women were also similar in the two subgroups. Thus, the clinical features of patients who had chronic hepatitis C could be compared with those of patients who had HCV and HGV infection without respect to interferon-α therapy (Table 1). Patients with HGV infection were significantly younger than those without HGV infection. Other clinical features did not differ between HGV RNA-positive and HGV RNA-negative patients (Table 1).

Hepatitis C virus genotype and serum HCV RNA level were compared between the 21 patients with HGV RNA and the 52 patients without HGV RNA who were randomly selected from the 168 HGV-negative persons. The HCV genotypes and HCV RNA levels were distributed similarly in the two groups (Table 2).

Effect of Hepatitis G Virus Infection on Response of Hepatitis C Virus to Interferon-α Therapy

Of the 101 patients receiving interferon-α, 36 had a sustained loss of HCV RNA and normalization of alanine aminotransferase levels; they were therefore considered to have responded to interferon-α. The remaining 65 patients were positive for HCV RNA 6 months after completing therapy and thus were classified as nonresponders. The rate of HCV response to interferon-α did not differ between patients with and those without HGV infection (Table 2).

Response of Hepatitis G Virus to Interferon-α

Response of HGV to interferon-α was analyzed in 9 of 10 patients with HGV infection. The serum HGV RNA level decreased during interferon-α therapy in all 9 patients. The geometric mean titer of HGV RNA just after interferon-α therapy (mean, 6.3 RNA copies/mL; range, 0.0 to 5000 RNA copies/mL) was significantly (P = 0.008; Wilcoxon rank-sum test) lower than the titer just before therapy (mean, 3200 RNA copies/mL; range, 20 to 1 000 000 RNA copies/mL). Two patients had a sustained loss of HGV RNA 6 months after discontinuation of therapy. In the remaining 7 patients, HGV RNA level increased after cessation of therapy to levels similar to those just before therapy. Thus, the sustained response rate of HGV (22%; 2 of 9 patients) was lower than but not significantly different from (P > 0.2; Fisher exact test) the sustained response rate of HCV (36%; 36 of 101 patients). However, a dichotomy was seen in the response to interferon-α; two patients who had a sustained loss of HGV RNA did not clear HCV RNA, and three patients who had a sustained loss of HCV RNA did not clear HGV RNA. In the latter three HCV responders, alanine aminotransferase levels remained normal despite the reappearance of HGV viremia.

Discussion

Approximately 10% of patients with chronic hepatitis C are also infected with HGV [5, 6]. Although the precise routes of HGV transmission have not been established, this agent is parenterally transmitted through blood transfusion and exposure to shared needles during injection drug use [4]. In our study, the frequency of previous blood transfusion was similar in patients with HGV and HCV co-infection and patients with hepatitis C alone. The apparent link between HGV and HCV infections probably reflects common exposures and transmission patterns rather than a specific interdependence of the two agents.

Patients with chronic hepatitis often harbor more than one hepatitis agent [4, 13-15], and important interactions between HBV and HCV have been documented [15]. The relation between HCV and HBV replication is reciprocal: Increasing replication of one agent can diminish the replication of the other. The key question underlying our study is whether coexistent HGV infection alters the level of viremia, clinical course, or treatment response of HCV infection. Although it is unclear whether our findings are applicable to non-Japanese persons, we found no evidence of such an effect on any of these variables. The following findings support this claim.

First, the HCV RNA level in patients with HGV and HCV infection was the same as in patients with HCV infection alone. In addition, Nakatsuji and colleagues [5] reported that the serum level of HGV RNA did not differ between these two groups of patients. These two studies showed no evidence of unidirectional or bidirectional viral interference between HGV and HCV.

Second, no evidence suggested that HGV infection increased the severity of hepatitis C. When patients with chronic hepatitis C were compared with those who had HCV and HGV infection, no differences were seen in the mean alanine aminotransferase level or liver histologic findings. Further, in five co-infected patients in whom HGV and HCV responses to interferon-α therapy were dissociated, hepatic inflammation after discontinuation of therapy seemed to depend on HCV replication, not on HGV replication. These data suggest that HGV has limited pathogenicity compared with HCV, and they are consistent with previous observations that HGV-positive blood donors were no more likely to have elevated alanine aminotransferase than were HGV-negative donors [4]. Several viral and host factors have been reported to influence the response of HCV to interferon-α [8, 16-18]. We found no effect of HGV on the HCV response; complete response to interferon-α, as assessed by the sustained loss of HCV RNA, was similar in both patient groups.

Finally, we studied the susceptibility of HGV to interferon-α therapy. The serum level of HGV RNA decreased during interferon-α therapy in all nine patients in whom this value was measured. However, as with HCV infection, viremia returned after discontinuation of therapy in most patients; only 2 of 9 patients (22%) had a sustained HGV response. Thus, sustained remission in HGV infection is of the same low order of magnitude as in HCV infection.

Drs. Alter, Nakatsuji, and Shih: Department of Transfusion Medicine, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892.

Dr. Kim: Genelabs Technologies, Inc., 505 Penobscot Drive, Redwood City, CA 94063.